The idea from this blog came from the "Extra-Curricular Topics" I teach in my MBA classes, a 10-minute interlude where I teach an academic paper with significant real-world relevance. This expanded to an opt-in Google Group where I wrote to former students summarizing an interesting paper that I come across in a seminar or conference, and from there came this blog. However, the first few blog posts ended up being on different topics - this is the first post on the intended theme.
One very interesting paper I saw presented at the LBS external seminar series is http://www.people.hbs.edu/cmalloy/pdffiles/malcolou.pdf by Lauren Cohen (HBS), Dong Lou (LSE), and Chris Malloy (HBS). It uses analyst conference calls to form a trading strategy. After Regulation FD, firms are no longer allowed to disclose information selectively to certain groups of investors and not others, so analyst conference calls are an important way in which they disseminate information. All analysts are allowed to participate.
However, what I didn’t know until I saw the talk, was that firms can choose which analysts they would like to speak and ask questions in the meeting. Analysts can signal if they would like to ask a question, but the firm has full discretion on which analysts to call upon. The paper shows that certain firms will selectively choose optimistic analysts (i.e. those who give them high ratings) and prevent pessimistic analysts from doing this. Such a strategy is bad in the long-run, because analysts that are not allowed to speak may end up dropping their coverage of the firm (and analyst coverage is useful for boosting stock liquidity). However, myopic firms who are focused on boosting the short-term stock price may engage in this strategy – indeed, these are firms that are just about to issue equity (so they want to prop up the short-term stock price) and end up announcing earnings that just meet the earnings forecast or beat it by 1 cent (suggesting they’ve done something myopic like cut R&D to meet the target).
The conference call information is public information which can be used to form a long-short strategy: sell the firms that engage in such manipulation and buy the firms that don’t. This strategy earns 95 basis points per month, nearly 12% per year, which is huge alpha. The manipulating firms also end up having negative earnings announcements in the future, having to restate previously announced earnings (implying that the previous earnings were falsified), and using discretionary accounting accruals to artificially boost earnings. This paper is a great example of using a clever institutional detail (the fact that it’s the firm who gets to decide who speaks on a conference call) to find an extremely profitable trading strategy.
You might think - shouldn't the SEC ban companies from being allowed to selectively pick and choose who speaks? Well, actually there's a good reason for allowing companies this discretion. It allows them to stop Bruce Wayne from Wayne Enterprises from calling in (see p13 of http://dealbreaker.com/uploads/2013/05/ARCHER_OL-Transcript-2013-05-30T12_001.pdf).
